Database support license = unfair trade practice… ??

The other day, I used the Oracle transportable tablespace mechanism at a customer site to rapidly copy a schema from one database to another. The customer DBA working with me stared in awe and said “This is slick! I have never used this new feature before...” Now it was my turn to be awe-struck. I gently reminded him that Oracle had released the transportable tablespace feature in Oracle8i back in 1999! And this was 2006 and we were working with Oracle10g… the transportable tablespace mechanism was not a new feature anymore – unless one has been hanging with Rip Van Winkle for the last 7 years.

That little incident made me wonder... Many of our customers rely on multiple database platforms, tools and applications. Some of these applications use older database versions. Often even the applications that use the newest database version pretend the database is an older version. In other words, the apps rarely use any newer feature/functionality from the latest releases. In such situations, I always wonder what advantage did the customer gain by upgrading their database version?

When I ask the customer this question, they say “Oh, we had no choice, the vendor has desupported the earlier version.” I naively ask “Were you using their support pretty regularly then and that’s what prompted you to upgrade since you were at risk of not being supported any more?”. Pat comes the answer from their DBA – “We were already on the terminal release for that version; it was stable in our environment, so we weren’t really calling their support line at all. But then, we were anyway paying them support fees, the upgrade didn’t cost us anything, so we went ahead and upgraded.”

An upgrade didn’t cost anything? Yeah right! When asked how long and how many resources it took them to upgrade, the answer is usually in several man-months to provision new machines, perform installation, configuration, testing of the newer database release and all the related apps to ensure those apps can function in the new release without any kind of degradation. So now I’m even more confounded. If you weren’t using the vendor’s support services, why would you care if you were de-supported or not? Then again comes the answer “oh, it’s our corporate policy that we can’t have un-supported software”. Policy made by who? Does this so-called policy-maker understand the ramifications of upgrading the database without a clear business need? Does this person really understand the ROI on the support fees your company has been paying the software vendor?

It seems somewhat of a catch-22. If you don’t upgrade your already stable release, then you are at risk of running on a de-supported release. So you spend lots of time and resources upgrading and then move to a newer release that may not be as stable as the prior release you were on. And on top of that, you barely use any (read zero!) of the newer features available in the new release. So besides having the peace of mind that you are running on a “supported” platform, what else do you gain? Nothing but the problems that are introduced with the newer release…

Oracle, Microsoft and IBM have all released newer versions of their databases in the last couple of years. They charge an average of 20% (an arm and a leg!) for support. I say, given that very few newer functionality is being utilized, customers should just stop paying for support. If there are bugs in the product, the vendor should provide a patch regardless of whether the customer has purchased support or not. After all, didn’t the customer pay for the software expecting it to work? If it doesn’t work, whose fault is it? Why does the vendor provide a solution to that bug only if the customer has paid an extra 20% for this so-called support?

Imagine what would happen if the average retailer operated like the average software company… You pick up a book at Borders and find that it’s missing a couple of pages. You take it back asking for another copy and they ask you if you have purchased a support license…

If this scenario is not acceptable in retail, why is it the norm in the software industry? Is it because in the former case, people are dealing with their “own money” whereas in the latter, it’s just their “company’s resources”??

I say, it’s time for companies to rally together and file a class action suit against the software monoliths asking for support fees to be refunded back to them. If the software is buggy, then the software vendor needs to get it fixed and supply a patch pronto regardless of whether the customer has a support license or not! After all, they have paid for that software expecting it to be functional. No sane person pays for anything expecting it to be buggy.

As far as support fees funding “future innovation” goes, newer releases/upgrades could be chargeable separately and customers at their discretion can look at the new release and see if it has value for them and if their apps would run on it, etc. and accordingly make the decision to upgrade (or not!). They shouldn’t have to upgrade because the older version is going to be “de-supported”. If the vendor is going to de-support an older release, the vendor should then provide the newer release at no additional charge to the customer that doesn’t have a business need for upgrading otherwise.

I don’t know if customers would be willing to go through the pain of litigation. But maybe they can just stop paying for support. Unless they put their foot down, the software vendors have no motivation to change and it’s business as usual all around where the customer shells out gobs of money for “support” that they sparingly use. The returns are just not there for the customer.

Software vendors should charge for the software or charge for the support; but charging for both given the above circumstance resorts to an unfair trade practice in my book. What am I missing?